Abstract Early detection is one of the keys to winning the war on cancer. Because current imaging and tissue biopsy based diagnosis are invasive, expensive, and in some cases dangerous, they are impractical as an early cancer screen, as well as therapeutic and recurrence monitoring. Liquid biopsy is an exciting new form of biopsy that can complement cancer detection assays from the analysis of bodily fluids such as blood or urine because it is non-invasive, easier, safer, and more cost effective. With regards to body fluids based cancer detection and monitoring, blood analysis is the most reliable source, with three types of analysis emerging as possible approaches for detection- circulating tumor cells (CTCs), cell free DNA (cfDNA), and extracellular vesicles (EVs). CTCs have been investigated by many researchers and companies in the past 30 years. Many technologies are available for their capture and identification. However, CTCs are rare, particularly in early stage cancer patients (<5 CTCs in 1 mL blood), making quantitative analysis difficult and less reliable. Many cancer patients also do not show CTCs in blood because of endothelial-to-mesenchymal-transition in circulation. Today, CTC based cancer diagnosis has not been widely used in clinic. Cell free DNA (cfDNA) has gained its success in noninvasive parental testing (NIPT). But the content of DNA targets from dead target cells such as early stage cancer is low, which requires PCR for amplification and Sequencing for identification. Consequently, the procedure is costly (>$1,000 per test in NIPT). cfDNA for early stage cancer detection is not proven, and many cancers are not initiated by DNA mutations. Extracellular vesicles (EVs) are abundant in blood (1012 ? 1014 in 1 mL blood), making quantitative analysis easier. Coding (e.g. mRNA) and non-coding (e.g. microRNA and lncRNA) RNAs encapsulated in EVs and membrane proteins on the surface of EVs can serve as good disease biomarkers. They also play important roles in disease metastasis and relapse, making the capture and characterization of individual EVs, not average information from all EVs in the sample as in the existing methods, highly valuable in disease diagnosis and medicine development. Many cancer types including pancreatic cancer have very high mortality rates because it is difficult to detect their presence in the early stage. Current biomarkers for pancreatic cancer such as serum CA-19-9 protein face high false positive and negative rates. The proposed studies seek to develop a robust and user-friendly novel molecular beacon in lipoplex nanoparticles for extracellular vesicles based cancer diagnosis and validate its clinic potential by detecting Glypican 1 mRNA and KRAS mutation in extracellular vesicles from pancreatic cancer patient serum samples for imminent use in the fields of early cancer diagnosis and relapse monitoring. The specific Phase I aims and milestones of this project are: Specific Aim 1: Design and testing of a toehold initiated (Ti) and catalyzed hairpin DNA circuit (CHDC) CLN-MB which are highly stable, specific and sensitive for GPC1 and KRAS mutation genes. Milestones: (i) Detection of GPC1 and KRAS mutation mRNAs in EVs secreted from <2x104 pancreatic cancer cells in 1 mL blood with >95% repeatability; (ii) lyophilized CLN-MB nanoparticles for >6 months shelf-life and <10% performance variation; (iii) <10% variation in100 assays using a conventional microscope as an analyzer. Specific Aim 2: Dual-site validation of EV GPC1/KRAS mRNA biomarker for early PDAC diagnosis. Milestones: (i) User-friendly assay with <4 hours assay time; (ii) <10% false positive/negative prediction from >200 pancreatic cancer patient samples at OSU and MSKCC.